利用通用智能体进行情境化时间序列分析
Harnessing Generalist Agents for Contextualized Time Series
AI总结：
提出TimeClaw框架，通过集成可执行时间工具、经验驱动能力进化和情景多模态记忆，使通用大语言模型智能体具备情境化时间推理能力，在能源、金融等多领域基准上取得性能提升。
摘要：
时间序列通常嵌入在丰富的上下文中，这对于整体建模至关重要。此外，现实世界的从业者通常需要用于分析时间动态的端到端工作流，其中广泛研究的任务（如预测）只是更广泛解决方案循环中的一个步骤。虽然通用AI智能体为复杂上下文下的此类工作流提供了有前景的接口，但它们主要运行在文本空间中，并未与结构化时间信号完全对齐。在这项工作中，我们引入了TimeClaw，一个用于时间序列的智能体框架，它为通用大语言模型智能体配备了情境化时间推理所需的时间序列原生运行时支持。TimeClaw集成了可执行的时间工具以进行有根据和可审计的分析，经验驱动的能力进化以创建可重用的分析例程，以及用于检索相关推理轨迹的情景多模态记忆。这些组件共同解锁了带有上下文信息的开放式时间推理。在涵盖能源、金融、天气、交通和其他现实世界领域的多个基准上的广泛评估表明，TimeClaw的性能得到了提升。代码可在https://github.com/iDEA-iSAIL-Lab-UIUC/TimeClaw获取。
英文摘要：
Time series are often embedded in rich contexts that are essential for holistic modeling. Moreover, real-world practitioners often require end-to-end workflows for analyzing temporal dynamics, where widely studied tasks such as forecasting are only one step in a broader solution loop. While generalist AI agents offer a promising interface for such workflows under complex contexts, they still operate primarily in textual spaces that are not fully aligned with structured temporal signals. In this work, we introduce TimeClaw, an agentic harness framework for time series that equips generalist LLM agents with the time series-native runtime support needed for contextualized temporal reasoning. TimeClaw integrates executable temporal tools for grounded and auditable analysis, experience-driven capability evolution for creating reusable analytical routines, and episodic multimodal memory for retrieving relevant reasoning traces. Together, these components unlock harnessed open-ended temporal reasoning with contextual information. Extensive evaluation on multiple benchmarks covering diverse tasks across energy, finance, weather, traffic, and other real-world domains demonstrates improved performance of TimeClaw. Code is available at https://github.com/iDEA-iSAIL-Lab-UIUC/TimeClaw.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
